Methods for optical character recognition (ocr)

Patent metadata

Field	Value
Publication number	US-2019102649-A1
Application number	US-201715720550-A
Country	US
Kind code	A1
Filing date	Sep 29, 2017
Priority date	Sep 29, 2017
Publication date	Apr 4, 2019
Grant date	—

Abstract

A method is provided for Optical Character Recognition (OCR). A plurality of OCR decoding results each having a plurality of positions is obtained from capturing and decoding a plurality of images of the same one or more OCR characters. A recognized character in each OCR decoding result is compared with the recognized character that occupies an identical position in each of the other OCR decoding results. A number of occurrences that each particular recognized character occupies the identical position in the plurality of OCR decoding results is calculated. An individual confidence score is assigned to each particular recognized character based on the number of occurrences, with a highest individual confidence score assigned to a particular recognized character having the greatest number of occurrences. Determining which particular recognized character has been assigned the highest individual confidence score determines which particular recognized character comprises a presumptively valid character for the identical position.

First claim

1 . A method for Optical Character Recognition (OCR), the method comprising:
obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more OCR characters, each OCR decoding result comprising a plurality of positions; comparing a recognized character in each OCR decoding result of the plurality of OCR decoding results with the recognized character that occupies an identical position in each of the other OCR decoding results; calculating a number of occurrences that each particular recognized character occupies the identical position in the plurality of OCR decoding results; assigning an individual confidence score to each particular recognized character based on the number of occurrences, with a highest individual confidence score assigned to a particular recognized character having the greatest number of occurrences; and determining which particular recognized character comprises a presumptively valid character for the identical position by determining which particular recognized character has been assigned the highest individual confidence score.

2 . The method according to claim 1 , further comprising repeating the comparing, calculating, assigning, and determining steps to determine the presumptively valid character for each position of a presumptively valid OCR decoding result.

3 . The method according to claim 1 , wherein obtaining the plurality of OCR decoding results comprises:
capturing the plurality of images of the same one or more OCR characters; and decoding each image after the plurality of images is captured or after each image of the plurality of images is captured.

4 . The method according to claim 2 , further comprising:
associating each OCR decoding result with a total confidence score based on a combination of the individual confidence scores assigned to each particular recognized character therein; identifying the OCR decoding result that is associated with the highest total confidence score; and selecting the OCR decoding result that is associated with the highest total confidence score as the presumptively valid OCR decoding result.

5 . The method according to claim 1 , further comprising, prior to the calculating step, transmitting the recognized character from capturing and decoding each image of the plurality of images to a memory space and aggregating each occurrence of the recognized character in the memory space for calculating the number of occurrences.

6 . The method according to claim 4 , wherein selecting the OCR decoding result comprises selecting the OCR decoding result as the presumptively valid decoding result even when the selected OCR decoding result does not match another OCR decoding result of the plurality of OCR decoding results.

7 . The method according to claim 4 , wherein selecting the OCR decoding result comprises selecting the OCR decoding result as the presumptively valid decoding result even when none of the plurality of decoding results is correctly decoded in its entirety.

8 . The method according to claim 2 , further comprising generating the presumptively valid decoding result from an ordered combination of the presumptively valid characters, the presumptively valid decoding result being absent from the plurality of decoding results.

9 . The method according to claim 1 , further comprising:
predetermining a lexicon for the plurality of OCR decoding results; and determining whether to include or exclude the recognized character in the calculation of the number of occurrences if the recognized character is not included in the predetermined lexicon.

10 . The method according to claim 1 , wherein the one or more OCR characters in each image of the plurality of images comprises greater than one OCR character and comprises a line of text.

11 . A method for Optical Character Recognition (OCR), the method comprising:
obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more characters, each OCR decoding result comprising a plurality of positions; comparing a first recognized character of each OCR decoding result of the plurality of OCR decoding results with the first recognized character of each of the other OCR decoding results to determine one or more first recognized characters; assigning each first recognized character of the one or more first recognized characters an individual confidence score according to its respective number of occurrences as the first recognized character in the plurality of OCR decoding results; and determining which particular first recognized character comprises a presumptively valid first recognized character by determining the particular first recognized character that has been assigned a highest individual confidence score.

12 . The method according to claim 11 , further comprising:
comparing each succeeding recognized character of each OCR decoding result with a recognized character occupying the identical position in each of the other OCR decoding results and assigning each succeeding recognized character the individual confidence score according to its respective number of occurrences, until comparison and assignment are completed over all recognized characters included in each position of each OCR decoding result of the plurality of OCR decoding results; and determining which particular recognized character comprises a presumptively valid character for each position by determining the particular recognized character for each position that has been assigned the highest individual confidence score.

13 . The method according to claim 12 , further comprising:
associating each OCR decoding result with a total confidence score based on a combination of the individual confidence scores assigned to each particular recognized character in each OCR decoding result; identifying the OCR decoding result that is associated with the highest total confidence score; and selecting the OCR decoding result that is associated with the highest total confidence score as a presumptively valid OCR decoding result.

14 . The method according to claim 13 , wherein selecting the OCR decoding result comprises selecting the OCR decoding result as the presumptively valid decoding result even when the selected OCR decoding result does not match another OCR decoding result of the plurality of OCR decoding results.

15 . The method according to claim 13 , wherein selecting the OCR decoding result comprises selecting the OCR decoding result as the presumptively valid decoding result even when none of the plurality of decoding results is correctly decoded in its entirety.

16 . The method according to claim 12 , further comprising generating the presumptively valid decoding result from an ordered combination of the presumptively valid characters for each position, the presumptively valid decoding result being absent from the plurality of decoding results.

17 . The method according to claim 11 , further comprising:
predetermining a lexicon for the plurality of OCR decoding results; and determining whether to include or exclude the recognized character in calculating its respective number of occurrences if the recognized character is not included in the predetermined lexicon.

18 . A method for Optical Character Recognition (OCR), the method comprising:
obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more characters, each OCR decoding result comprising a plurality of positions; associatin

Assignees

• Datamax Oneil Corp

Inventors

• Ackley H Sprague

Classifications

• G06V30/192 Primary

  using simultaneous comparisons or correlations of the image signals with a plurality of references · CPC title

• G06V10/809

  of classification results, e.g. where the classifiers operate on the same input data · CPC title

• G06F18/254

  of classification results, e.g. of results related to same input data · CPC title

• G06V30/10

  Character recognition · CPC title

• G06K2209/01

  Physics · mapped topic